Tool Use on Aggregate Performance
76.8Avg1 ScoreNRE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| NREModel=Kimi-k2.62026.07 | 76.8 | 79.9 | |
| VanillaModel=Opus 4.72026.07 | 76.7 | 78.1 | |
| NREModel=Haiku 4.52026.07 | 75.9 | 79 | |
| NREModel=Mistral 3.52026.07 | 75 | 77.5 | |
| VanillaModel=Sonnet 4.62026.07 | 74.5 | 76.2 | |
| NREModel=gpt-5.4-mini2026.07 | 72.9 | 76 | |
| NREModel=gpt-oss-120b2026.07 | 72.7 | 77 | |
| NREModel=Deepseek-v32026.07 | 72.4 | 75.2 | |
| VanillaModel=Kimi-k2.62026.07 | 72.3 | 73.2 | |
| GEPAModel=Kimi-k2.62026.07 | 71.2 | 74.1 | |
| VanillaModel=Haiku 4.52026.07 | 59.2 | 61.1 | |
| RLMModel=Kimi-k2.62026.07 | 57.8 | 59.8 | |
| GEPAModel=gpt-oss-120b2026.07 | 56.6 | 59.4 | |
| VanillaModel=Deepseek-v32026.07 | 54.7 | 57.2 | |
| VanillaModel=gpt-oss-120b2026.07 | 51.2 | 52.7 | |
| VanillaModel=Deepseek-R12026.07 | 38.4 | 41.5 | |
| GEPAModel=Deepseek-v32026.07 | 35.7 | 38.2 | |
| VanillaModel=gpt-5.4-mini2026.07 | 34.9 | 37.9 | |
| RLMModel=gpt-oss-120b2026.07 | 22.1 | 24.2 | |
| RLMModel=Deepseek-v32026.07 | 1.6 | 2.4 |